Metraipolis › archive › dc_a229131dc996
Neural Expansion: A Unified Mechanism for How Deep Neural Networks Generalize — arXiv:2610.23716
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.23716
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T08:43:24.431Z
- SHA-256 of the archived text
de11d76c80310ce643b6142bd1b85d6bb47edb96d04dee3752242703339c7698- Stable URL of this page
https://metraipolis.com/documento/dc_a229131dc996
How to cite
Neural Expansion: A Unified Mechanism for How Deep Neural Networks Generalize — arXiv:2610.23716 — arXiv (en), https://arxiv.org/abs/2610.23716 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a229131dc996 (Metraipolis archive dc_a229131dc996, sha256 de11d76c80310ce643b6142bd1b85d6bb47edb96d04dee3752242703339c7698)
BibTeX
@misc{dca229131dc996,
title = {Neural Expansion: A Unified Mechanism for How Deep Neural Networks Generalize — arXiv:2610.23716},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.23716},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a229131dc996; sha256 de11d76c80310ce643b6142bd1b85d6bb47edb96d04dee3752242703339c7698},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23716. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Deep neural networks demonstrate remarkable generalization despite being highly over-parameterized, challenging traditional statistical learning theories. While several theoretical paradigms such as margin-based bounds, PAC-Bayesian analyses, and neural tangent kernels provide global explanations, they often lack mechanistic insight into how generalization emerges from training dynamics. We introduce Neural Expansion, a unified geometric framework that characterizes generalization through the expansion of directionally-selective ray coverage in the input space. Using Generalization Intervals (GIs), we quantify direction-specific regions around training samples where the model's predictions remain stable. Our extensive empirical studies across diverse architectures and datasets reveal that training leads to progressive expansion of ray coverage, particularly along directions with low Jacobian singular values. This mechanism helps explain a large fraction of correctly classified test inputs, adversarial examples, and even mislabeled or out-of-distribution samples through the geometry of ray coverage rather than instance-level memorization. By connecting input-space curvature to predictive behavior, our framework provides a mechanistic foundation that complements existing global theories of deep learning. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.23716 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-19T08:39:10.165Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).